Assessment of Flood Vulnerability in the Upper Benue River Basin, Nigeria, Using the Fuzzy Analytic Hierarchy Process (Fuzzy AHP) Technique

Abstract

Yakubu Stephen Zirankuthii, Emeka Daniel Oruonye, Adelalu Temitope Gabriel and Musa Hassan

Flooding remains one of the most recurrent and devastating natural hazards in Nigeria, particularly within the Upper Benue River Basin (UBRB), where rapid urbanization, deforestation, and intense rainfall events have heightened exposure and vulnerability. This study applied the Fuzzy Analytic Hierarchy Process (Fuzzy AHP) integrated with Geographic Information Systems (GIS) to assess spatial flood vulnerability across the basin. Eleven flood-conditioning factors; rainfall, slope, drainage density, topographic wetness index, elevation, land use/land cover, soil type, distance from rivers, Modified Normalized Difference Water Index (MNDWI), Normalized Difference Vegetation Index (NDVI), and terrain ruggedness, were analyzed. Expert judgments were converted into fuzzy pairwise comparison matrices, and weights were derived through Chang's extent analysis method. Results showed that the rainfall (0.175), drainage density (0.068), and slope (0.066) as the most influential factors. The consistency ratio (0.0398) indicated high logical coherence, while sensitivity analysis confirmed model robustness with less than 5% variation in the ranking under ±10% weight perturbation.

The resulting vulnerability map classified the basin into five zones. Very high vulnerability zones were concentrated along the Benue, Gongola, and Taraba Rivers, affecting Yola, Jimeta, Jalingo (particularly its low-lying sections), Girei, Numan, and Toungo. High vulnerability extended to Gombe Town, Kaltungo, Lau, and Karim Lamido. Medium vulnerability was observed in Akko, Yamaltu-Deba, Zing, and Yorro. Low and very low vulnerability zones were largely confined to the Mambilla Plateau, Gotel Mountains, and the highlands of Gashaka-Gumti National Park (Gembu, Nguroje, and Maisamari) The Fuzzy AHP-GIS approach effectively captured uncertainty in expert judgments, providing a reliable framework for flood-vulnerability assessment in data-scarce environments such as the UBRB.